Scientific Investigations Report 2008–5059
U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2008–5059
Back to Table of Contents
There has always been a need for reliable information on the configuration of the water table in the Portland, Oregon area (fig. 1), for a multitude of purposes. This information typically is used in construction of buildings, roads, and infrastructure; well drilling; evaluation of aquifer susceptibility; and the design of monitoring programs to determine the extent and severity of possible aquifer contamination. However, recent and planned construction of new infrastructures for diverting stormwater runoff have raised concerns regarding the protection of ground-water resources in the Portland area and have emphasized the need for information about the configuration of the water table. Presently (2008), large numbers of underground injection control (UIC) systems (for example, stormwater injection systems, sumps, and drywells) are used to divert stormwater runoff into the subsurface. Additional UIC systems are expected to be constructed, and stormwater runoff also will be diverted to other types of stormwater drainage systems such as vegetated swales, pervious pavement, disconnected downspouts, and other diversion methods that are designed to allow for the infiltration of stormwater. The goals are to manage surface runoff and to protect the quality of water entering rivers, streams, and the ground-water flow system. However, information is needed regarding the configuration of the water table to help determine the appropriate use of these diversion methods to meet regulatory requirements and to minimize the effect on ground-water quality and ground-water levels. To help provide this information, the U.S. Geological Survey (USGS), in cooperation with the City of Portland, City of Gresham, Clackamas County’s Water Environment Services, and Multnomah County began a study in 2003 to determine the configuration of the water table in the Portland area.
The purpose of this report is to present estimated depth-to-water and water-table elevation maps for the Portland metropolitan area, along with estimates of the relative uncertainty, and seasonal water-table fluctuations. The study area comprises most of the Portland metropolitan area, where an increasing need for information on the position of the water table and issues pertaining to UIC systems have come to the forefront. The area previously was the focus of an extensive USGS regional ground-water study providing a foundation for the current investigation.
The method of analysis used to determine the configuration of the water table in the Portland area relied on water-level data collected from shallow wells as part of the current study or used in previous USGS investigations and on surface-water features that are representative of the water table, including major rivers, streams, lakes, wetlands, and springs. Depth to water and water-table elevation were interpolated independently based on kriging methods and then combined into a single representation of the water-table position. Kriging methods also were used to develop a map of relative uncertainty associated with the values of the water-table position. Range of seasonal water-table fluctuations was evaluated based on wells with multiple water-level measurements distributed throughout the seasons and then categorized by hydrogeologic unit.
This section contains a short introduction to some ground-water terms and principles to help the reader better understand concepts discussed in the main body of the report.
Technical definition of the water table is the imaginary surface in an unconfined aquifer at which the pressure is atmospheric (Lohman, 1972b, p. 1). The water table is thought of as the surface representing the top of the saturated zone, below which all pores in the rock matrix are filled with water (fig. 2). The water table is defined by the levels at which water stands in wells that just penetrate the top of the water body (Lohman, 1972a, p. 14). Additional information on the position of the water table is provided by the locations of surface-water features that interact with the water table, which can include rivers, streams, lakes, wetlands, and springs (Fetter, 1994, p. 114-115).
The above definition of the water table excludes
The two conventions typically used to define the position of the water table (fig. 2) are depth and elevation: (1) depth to the water table below land surface is referred to as “depth to water” in this study and also referred to as the “unsaturated zone thickness;” and (2) the elevation of the water table above a datum (NAVD 88, a representation of mean sea level is used in this study; see section “Datums” for more information), which will be referred to as “water-table elevation.” The two conventions are related and can be equated as follows:
water-table elevation = land-surface elevation – depth to water below land surface (elevations are in feet NAVD 88; depth is in feet).
Depth-to-water values can be transformed (converted) to values of water-table elevation based on this equation, if the land-surface elevation is known and, conversely, the water-table elevation can be transformed to depth to water. The terms “water-table position” or “water-table level” will be used to refer to the vertical location of the water table independent of the convention used to define it.
The water table generally is a subdued replica of land surface (Latham, 1878, p. 207-208; King, 1892, p. 15 and 18; King, 1899, p. 99; Russell, 1963, p. 10). This condition more commonly is observed in humid areas with relatively thin unsaturated zones. The configuration (shape or form of the surface) of the water table is a function of the geometry of the land surface, rate and location of ground-water recharge and discharge, aquifer properties, and extent, thickness, and shape of the aquifer and adjacent confining units (Haitjema and Mitchell-Bruker, 2005, p. 784). These factors generally do not change with time, with the exception of recharge and discharge and, therefore, variations in the water-table configuration represent changes in recharge, discharge, or both. Changes in the rates of recharge or discharge cause changes in ground-water storage, which are represented by water-table fluctuations. The water table rises due to increased ground-water storage when the rate of recharge exceeds the rate of discharge and declines when these conditions are reversed (Veeger and Johnston, 1996, p. 28). Fluctuations can be the result of any influence that can change the amount of or location of recharge or discharge. Changes in recharge can result from variations in natural factors, such as in precipitation patterns or rates, or from human-induced modifications such as changes in impervious area, irrigation, septic systems, stream withdrawals, or stormwater runoff into UIC systems. Changes in discharge occur primarily as a result of variations in pumping rates, evapotranspiration, or changes in ground-water storage.
The quantity and timing of precipitation typically are the greatest influences on water-table fluctuations in the Portland Basin. Ground-water levels rise following precipitation of sufficient intensity and duration to satisfy surface runoff, evapotranspiration, and soil moisture deficits, with the residual resulting in recharge. Water levels decline after extensive periods of little or no precipitation as ground water discharges to rivers, streams, springs, or is utilized by pumping or evapotranspiration. Water-table fluctuations can occur on a wide range of time scales, from hours, as in the response to high intensity precipitation events or in response to changes in stream stage, to decades, as in response to long-term changes in climate or land use (Hogenson and Foxworthy, 1965, p. 12; McFarland and Morgan, 1996, p. 38; Conlon and others, 2005, p. 61). As a result, the water-table position is dynamic and constantly changing in response to changes in natural and human-induced stresses acting over a wide range of time scales.
The direction of ground-water flow in an unconfined aquifer is dependent on the hydraulic head, an indicator of the total energy available to move ground water through an aquifer; the hydraulic head at the water table is equal to the water-table elevation (Freeze and Cherry, 1979, p. 39; Taylor, and Alley, 2002, p. 3). Ground water in an unconfined aquifer flows downgradient from areas of high hydraulic head to areas of low hydraulic head. Therefore, a map depicting the configuration of the water-table elevation can be used to infer the direction of ground-water flow. However, ground-water flow is three dimensional and consists of vertical as well as horizontal components of flow. Water-table elevation maps are useful in determining the approximate horizontal direction of ground-water flow at the water table but should be used with caution and the knowledge that a vertical component of flow also is present and that the direction of lateral flow may vary with depth below the water table. A helpful description of ground-water level maps and how the maps may be used is provided by Bexfield (2002, p. 48-49).
Water-table elevation maps also are useful for identifying ground-water flow systems. A flow system consists of water within a bounded area that enters an aquifer at a recharge area, moves through the aquifer along pathways (flowpaths), and exits at a discharge area. Ridges in the water table usually represent ground-water flow divides from which ground water moves away in both directions normal to the ridgeline (U.S. Water Resources Council, 1980) and form boundaries between flow systems. A common classification scheme divides flow systems by relative size into local, intermediate, and regional systems that are differentiated primarily based on the distance between the recharge and discharge areas (Tóth, 1963, p. 4806). Local-flow systems are characterized by relatively short and shallow flowpaths that extend from a topographic high in the water-table elevation (recharge area) to an adjacent topographic low (discharge area). Intermediate flow systems include one or more local flow systems between respective points of recharge and discharge, and contain flowpaths that are longer and deeper than local flowpaths. Regional (or deep) flow systems may include multiple intermediate and local systems and have the longest and deepest flowpaths that begin at major ground-water divides and terminate at regional discharge areas, such as the Columbia and Willamette Rivers. Recognition of ground-water divides from water-table elevation maps can be useful in helping to identify and differentiate recharge and discharge areas, as well as the presence of local, intermediate, and regional ground-water flow systems.
Information describing the water-table position is useful for many applications, including understanding aquifer susceptibility, ground-water flow direction, contributing areas to wells, depth to water, or at what level ground water may affect construction activities. Representations of the water-table position include depth to water (equivalent to the thickness of the unsaturated zone) and water-table elevation. The primary focus of this study was to determine the water-table position in order to better understand the susceptibility of ground-water resources in the Portland area. The thickness of the unsaturated zone often is used as one of the components to evaluate aquifer susceptibility to contamination by estimating travel time or the potential for natural remediation in the unsaturated materials. This type of analysis commonly is used for point source contamination such as spills at point of use or transportation corridors, or for evaluating features such as leaking underground storage tanks and stormwater injection systems including UIC systems (Oregon Department of Environmental Quality, 2005a, p. 3, 7, 11, 36-37; 2005b, p. 3-4, 8-10, 20, 22-23, 27-31). Current estimates of the number of UIC systems in the Portland metropolitan area include about 11,000 publicly owned and an estimated 25,000 to 35,000 privately owned UIC systems. The distribution of UIC systems owned and maintained by Multnomah and Clackamas Counties and the cities of Portland, Gresham, Milwaukie, and Troutdale is shown in figure 3.
Information describing the depth to water has various other possible uses, such as well drilling or construction design (for example, roads, buildings, and sewers). Depth-to-water information also can be useful for nonpoint source issues such as the residential or agricultural application of pesticides and fertilizers. Information on water-table elevations is useful for determining the direction of shallow ground-water flow, which can be used for monitoring possible nearby downgradient effects from an area of concern or identifying nearby upgradient sources from areas of interest. Information on water-table fluctuations helps to describe the range of possible seasonal variation that might be expected. Results of the study also can be used (1) as a baseline to identify changes in water levels resulting from changes in natural or human-induced causes, (2) as a tool to help identify areas where more detailed studies and supplemental data are needed to provide for greater resolution, or (3) to constrain the water-table position and reduce the uncertainty in the estimate of the water-table position.
Ground-water hydrology of the Portland area has been the focus of numerous studies, although many of these studies are limited in scope. The most notable early investigations were those by Hogenson and Foxworthy (1965), who studied the ground-water resources for much of western Multnomah County; and Leonard and Collins (1983), who studied ground water in northern Clackamas County. A regional ground-water study by the USGS in the late 1980s and early 1990s represents the most recent extensive work done in the Portland Basin. Many reports resulted from this work, including:
The interested reader is referred to these reports for more detailed information on the hydrogeology of the Portland Basin.
The primary study area, consisting of the Oregon part of the Portland Basin in northwestern Oregon (fig. 1), covers about 615 mi2 and includes parts of Multnomah, Clackamas, Columbia, and Washington Counties. The Oregon part of the Portland Basin is defined here as the area bounded by the Tualatin Mountains to the west, the Columbia River to the north, the foothills of the Cascade Range to the east, and the Clackamas River to the south. The final boundary used for the presentation of the results is based on an uncertainty analysis and is limited to areas where the relative uncertainty of the water-table position was deemed to be acceptable. The final area of analysis excludes some parts of the primary study area, especially to the northwest in Columbia County, and extends beyond the bounds of the primary study area in other areas, depending on the quantity and distribution of available water-level data. The study area, which includes Portland, the largest city in Oregon, currently (2008) has a population of about 1 million. However, many of the suburban areas in Portland are rapidly growing, especially in Clackamas, Multnomah, and Washington Counties, contributing to an increasing population.
The Columbia River is the major drainage for the Portland Basin and flows westward out of the Columbia River Gorge until it reaches the confluence with the Willamette River, then flows northward. Topography in the study area includes flat areas, terraces, and hills. Land-surface elevations for the relatively flat areas along the floodplains of the major rivers range from 11 to 50 ft NAVD 88 (fig. 1). Rolling terraces occur above the floodplains and are remnant deposits of a series of ancient catastrophic floods (known as the Missoula Floods) (Waitt, 1985), with land-surface elevations that generally range from 50 to as high as 400 ft. Well-dissected uplands and hills are present above the terraces, with elevations more than 400 ft and sometimes exceeding several thousand feet, especially toward the foothills of the Cascade Range to the east (Trimble, 1963, p. 5 and 59).
Taylor and Hannan (1999, p. 50) describe the climate of the Portland area as relatively mild throughout the year, and characterized by cool, wet winters and warm, dry summers. Annual precipitation at the Portland International Airport (National Climatic Data Center Coop ID 356751) averages 37.1 in. (30-year normal for 1971–2000); however, precipitation generally is greater throughout most of the study area. About 50 percent of the precipitation occurs from December through February, with lesser amounts in spring and autumn, and very little during summer (Taylor and Hannan, 1999, p. 50). The Cascade Range is an orographic barrier to ocean weather systems moving across Oregon from west to east. As a result, precipitation generally increases eastward in the study area toward the foothills of the Cascade Range.
Overviews of the geology and hydrology of the Portland Basin presented in the following sections summarize more detailed descriptions in reports by (1) Swanson and others (1993), who discuss the thickness, extent, and lithology of hydrogeologic units in the basin; (2) McFarland and Morgan (1996), who describe the ground-water flow system of the basin, including its boundaries, hydraulic characteristics, and components of recharge and discharge; and (3) Morgan and McFarland (1996), who discuss the geology and hydrology related to the simulation of the ground-water flow system based on numerical modeling.
The northwest-trending Portland Basin was formed by structural deformation of the underlying Eocene and Miocene volcanic and marine sedimentary rocks. Late Miocene and younger fluvial and lacustrine sediments are overlain by unconsolidated Pleistocene Missoula Flood deposits and Holocene Columbia River alluvium (Swanson and others, 1993, p. 9; McFarland and Morgan, 1996, p. 6). The consolidated and unconsolidated basin-fill sediments are thickest adjacent to the Columbia and Willamette Rivers, where the thickness of the sediments may be as much as 1,800 ft. Hydrogeologic units in the Portland Basin, as defined by McFarland and Morgan (1996, p. 9), include, from youngest to oldest, the unconsolidated sedimentary aquifer, Troutdale gravel aquifer, confining unit 1, Troutdale sandstone aquifer, confining unit 2, sand and gravel aquifer, and older rocks. An additional unit, the undifferentiated fine-grained sediments, is mapped where the Troutdale sandstone aquifer is missing and confining units 1 and 2 cannot be differentiated. Results from several hydrogeologic units were combined as modified by Snyder and others (1998, p. 5) for the purpose of simplifying discussion and analysis. References to the undifferentiated fine-grained sediments in the remainder of the report will include confining units 1 and 2 (fig. 4).
Low-lying areas between the cities of Portland and Gresham, consisting of the floodplains of the Columbia and Willamette Rivers and the terraced areas above the floodplains, are of particular interest because of the large population and the great number and density of existing and proposed UIC systems. Primary surficial aquifers in this area consist of the unconsolidated sedimentary aquifer and the Troutdale gravel aquifer where the Troutdale aquifer is exposed at the surface or where it underlies the unconsolidated sedimentary aquifer in areas where that aquifer is nearly or completely unsaturated (Morgan and McFarland, 1996, p. 1, 57–58). These aquifers supplied more than 80 percent of the pumpage in 1987–88 (Collins and Broad, 1993, p. 8). The extent, thickness, distribution, size, and sorting of the materials composing these aquifers exert a strong influence on the position and character of the water table across the Portland area.
Throughout much of the low-lying areas in Portland (particularly to the north and west), the water table lies within the unconsolidated sedimentary aquifer (Hogenson and Foxworthy, 1965, p. 42). Materials in this part of the unconsolidated sedimentary aquifer consist of recent alluvium and repeated cycles of sediments grading upward from boulders, cobbles, and gravels to sands and silts. The degree of sorting of these cycles of layered materials is variable and though bedding (planes separating distinct sediments or rocks) is present there are many unconformities as the layers can be vertically and laterally discontinuous (Trimble, 1963, p. 58-62; Hogenson and Foxworthy, 1965, p. 26; Waitt and O’Connor, 2000, p. 7). These materials are a result of deposits from a minimum of 40 prodigious floods (collectively referred to as the Missoula Floods) (Bretz, 1969; Waitt, 1985, p. 1284). The Missoula Floods occurred about 12,700–15,300 years ago and were caused by the periodic catastrophic failure of the ice damming glacial Lake Missoula in western Montana releasing scores of colossal jökulhlaups (glacier-outburst floods) (Waitt, 1985, p. 1284). Floodwaters traveled hundreds of miles westward, carrying large quantities of rock material, which subsequently was deposited as layers of silt, sand, gravel, cobbles, and boulders in the Portland Basin after rapid loss in transporting power as the floodwaters moved from the high energy environment of the narrow Columbia River Gorge into the relatively low energy environment of the Portland Basin, where the area widens. Peak discharge for some floods exceeded 700 million ft3/s (Minervini and others, 2003) or about 3,600 times the mean daily discharge of the present day Columbia River at Vancouver, Washington. A hydraulic constriction of the Columbia River downstream of Portland prevented the rapid draining of the basin, resulting in a large lake extending about 115 mi down the Willamette Valley with a width exceeding 25 mi in places, and inundating the Portland area to depths as great as 400 ft (Trimble, 1963, p. 64–66; Minervini and others, 2003). Many of the familiar geomorphic features of the Portland area were either created or influenced by erosion or deposition associated with the episodes of flooding and are exhibited as ridges, terraces, channels, or depressions. These include well-known features such as Sullivan Gulch (west terminus of Interstate Highway 84) and Alameda Ridge (Allison, 1978a, p. 182). The thickness of the unconsolidated sedimentary aquifer typically is between 50 and 100 ft, with local accumulations of greater than 250 ft. The unconsolidated sedimentary aquifer is the most permeable of the aquifers in the Portland Basin; however, permeability varies substantially due to the high degree of heterogeneity of the aquifer materials, which can result in some local areas of perched ground water (McFarland and Morgan, 1996, p. 16 and 20).
The Troutdale gravel aquifer has been assigned by Swanson and others (1993, p. 27–29) to include several geologic formations—poorly to moderately cemented conglomerates of the Troutdale Formation, the volcaniclastic conglomerate of the Springwater, Walters Hill, and Gresham Formations, as well as locally thick accumulations of the Boring Lava. Conglomerates of the Troutdale Formation that extend over most of the study area were deposited by the ancestral Columbia River. The volcaniclastic conglomerate was derived from the Cascade Range and is present throughout much of the southeastern part of the study area. Finally, many of the isolated hills in the central and southern parts of the study area, including Mount Sylvania, Mount Tabor, Rocky Butte, Kelly Butte, Powell Butte, Mount Scott, and the Boring Hills, consist entirely of or contain a core of the Boring Lava, which can interlayer with the other units that form the Troutdale gravel aquifer (Trimble, 1963, p. 37; Swanson and others, 1993, p. 27). Thickness of the Troutdale gravel aquifer ranges from 100 to 400 ft throughout much of the basin (Swanson and others, 1993, p. 28). The Troutdale gravel aquifer has a relatively high permeability, although it also exhibits a large range in values (McFarland and Morgan, 1996, p. 16 and 20). Perched ground water occurs in some parts of the Troutdale gravel aquifer, particularly those associated with the Boring Lavas or the volcaniclastic conglomerates; however, the perched ground-water bodies are discontinuous and irregularly distributed (Hogenson and Foxworthy, 1965, p. 23, 25, 36, and 39).
Recharge to the Portland Basin from precipitation, runoff into UIC systems, and on-site waste-disposal systems (septic systems) was estimated by Snyder and others (1994, p. 30) over the entire Portland Basin and ranged from 0 to 49 in/ yr with a mean of 22 in/yr. Recharge primarily is through the infiltration of precipitation, although in areas of urban development, recharge from runoff into UIC systems and on-site waste-disposal systems contributed about 38 and 17 percent of the total, respectively. However, the number of UIC systems has increased substantially since 1994, although many on-site waste-disposal systems have since been removed as part of an extensive sewer project in mid-Multnomah County. Irrigation-return flow and losing streams may constitute locally important sources of seasonal recharge, but are believed to be insignificant on a regional scale. Ground-water recharge typically moves downward through the unsaturated zone to the water table (however, the unsaturated zone may be missing in areas where the water table is present at land surface). Ground water then may flow into unconfined, perched, or confined aquifers depending on the permeability of the aquifer and the overlying and underlying layers.
Ground-water movement primarily is controlled by the topography of the basin, which creates regional, intermediate, and local ground-water flow systems. The Columbia, Willamette, and Clackamas Rivers represent the regional discharge areas for the ground-water flow system in the study area (McFarland and Morgan, 1996, p. 20–23). Much of the ground water discharging to the rivers enters the ground-water system in upland recharge areas along the western Cascade Range, the Boring Hills, or the Tualatin Mountains; moves downward and horizontally toward the rivers; and finally moves upward to discharge to the rivers (Morgan and McFarland, 1996, p. 8; Hinkle and Snyder, 1997, p. 19–20, pl. 1; Snyder and others, 1998, p. 13–23 and unpub. data, 1998). An example of a discharge area for an intermediate ground-water flow system includes much of the Sandy River. Local ground-water flow systems are much smaller, with distances of only hundreds of feet between recharge and discharge areas (Morgan and McFarland, 1996, p. 9). Directions of ground-water flow can vary between each type of flow system and even within a flow system due to complexities in the flow system. This can present situations where the direction of ground-water flow at or near the water table may not be representative of the directions of ground-water flow at greater depths. These conditions have been illustrated in particle-tracking examples in the ground-water system of the Portland Basin (Hinkle and Snyder, 1997, p. 20, pl. 1).
Ground-water discharge in the Portland Basin is primarily to gaining streams and rivers, wells, and springs (McFarland and Morgan, 1996, p. 23). The largest component of ground-water discharge in the Portland Basin is to streams and rivers. Ground-water withdrawals from wells in the study area primarily are used for industry and public supply, with smaller amounts used for irrigation and domestic purposes (Collins and Broad, 1993, p. 7). Major springs in the study area are near Crystal Springs Creek and along the southern side of the Columbia River between the cities of Troutdale and Wood Village.
Back to Table of Contents